Improper Access Control Applied to Mirrored or Aliased Memory Regions

Aliased or mirrored memory regions in hardware designs may have inconsistent read/write permissions enforced by the hardware. A possible result is that an untrusted agent is blocked from accessing a memory region but is not blocked from accessing the corresponding aliased memory region.


Description

Hardware product designs often need to implement memory protection features that enable privileged software to define isolated memory regions and access control (read/write) policies. Isolated memory regions can be defined on different memory spaces in a design (e.g. system physical address, virtual address, memory mapped IO).

Each memory cell should be mapped and assigned a system address that the core software can use to read/write to that memory. It is possible to map the same memory cell to multiple system addresses such that read/write to any of the aliased system addresses would be decoded to the same memory cell.

This is commonly done in hardware designs for redundancy and simplifying address decoding logic. If one of the memory regions is corrupted or faulty, then that hardware can switch to using the data in the mirrored memory region. Memory aliases can also be created in the system address map if the address decoder unit ignores higher order address bits when mapping a smaller address region into the full system address.

A common security weakness that can exist in such memory mapping is that aliased memory regions could have different read/write access protections enforced by the hardware such that an untrusted agent is blocked from accessing a memory address but is not blocked from accessing the corresponding aliased memory address. Such inconsistency can then be used to bypass the access protection of the primary memory block and read or modify the protected memory.

An untrusted agent could also possibly create memory aliases in the system address map for malicious purposes if it is able to change the mapping of an address region or modify memory region sizes.

Demonstrations

The following examples help to illustrate the nature of this weakness and describe methods or techniques which can be used to mitigate the risk.

Note that the examples here are by no means exhaustive and any given weakness may have many subtle varieties, each of which may require different detection methods or runtime controls.

Example One

In a System-on-a-Chip (SoC) design the system fabric uses 16 bit addresses. An IP unit (Unit_A) has 4 kilobyte of internal memory which is mapped into a 16 kilobyte address range in the system fabric address map.

System Address

Mapped to

0x0000 – 0x3FFF

Unit_A registers : 0x0000 – 0x0FFF

0x4000 – 0xFFFF

Other IPs & Memory

To protect the register controls in Unit_A unprivileged software is blocked from accessing addresses between 0x0000 – 0x0FFF.

The address decoder of Unit_A masks off the higher order address bits and decodes only the lower 12 bits for computing the offset into the 4 kilobyte internal memory space.

In this design  the aliased memory address ranges are these:
0x0000 – 0x0FFF0x1000 – 0x1FFF0x2000 – 0x2FFF0x3000 – 0x3FFFThe same register can be accessed using four different addresses: 0x0000, 0x1000, 0x2000, 0x3000.The system address filter only blocks access to range 0x0000 - 0x0FFF and does not block access to the aliased addresses in 0x1000 - 0x3FFF range. Thus, untrusted software can leverage the aliased memory addresses to bypass the memory protection.
In this design the aliased memory addresses (0x1000 - 0x3FFF) could be blocked from all system software access since they are not used by software.Alternately, the MPU logic can be changed to apply the memory protection policies to the full address range mapped to Unit_A (0x0000 - 0x3FFF).

See Also

Memory and Storage Issues

Weaknesses in this category are typically associated with memory (e.g., DRAM, SRAM) and storage technologies (e.g., NAND Flash, OTP, EEPROM, and eMMC).

Comprehensive CWE Dictionary

This view (slice) covers all the elements in CWE.

Weaknesses without Software Fault Patterns

CWE identifiers in this view are weaknesses that do not have associated Software Fault Patterns (SFPs), as covered by the CWE-888 view. As such, they represent gaps in...

Weaknesses Introduced During Implementation

This view (slice) lists weaknesses that can be introduced during implementation.


Common Weakness Enumeration content on this website is copyright of The MITRE Corporation unless otherwise specified. Use of the Common Weakness Enumeration and the associated references on this website are subject to the Terms of Use as specified by The MITRE Corporation.